Information processing device, input control method and program
The invention provides a method which can contribute to an improvement in KPIs even in the entire time series range through an input stage to an AI engine and its output stage. As a preferred embodiment of the invention, an information processing device using an AI engine is provided, which includes an arithmetic section and a storage section. In the information processing device, the arithmetic section executes an important KPI reception unit which receives important KPI information designated by a user, and a variable selection unit which selects an input variable input to the AI engine, on the basis of the important KPI information.
1 . An information processing device using an AI engine, comprising:
a processor; and
a storage section,
wherein the processor is configured to:
receive important KPI information designated by a user,
select an input variable to be input to the AI engine, based on the important KPI information,
wherein the AI engine has a cause and an item as probability variables and is represented by a network defining conditional dependency between the cause and the item,
wherein the processor is configured to:
store in association with each other, in the storage section, a combination of a plurality of input conditions and a plurality of input procedure generation methods which are rules for ordering the items,
specify the plurality input conditions associated with the important KPI information,
specify a plurality of the items associated with the priority KPI information,
select the input procedure generation method based on the input conditions,
generate input procedure information for determining an order of the specified items based on the input procedure generation method,
transmit the input procedure information via an external terminal,
receive an item corresponding to the transmission and an answer to the item via the external terminal, and
calculate a probability of an output variable that is the cause using the AI engine with the item and the answer to the item as input variables of the AI engine.
2 . The information processing device according to claim 1 ,
wherein the processor is configured to specify a plurality of the items linked to the input, condition.
3 . The information processing device according to claim 2 ,
wherein the processor is configured to select, as the input procedure generation method one of:
a cause identification-emphasized input procedure generation method of assigning a high score to an item which more increases the probability of a high-probability cause when an input corresponding to the item is obtained, and
a cause isolation-emphasized input procedure generation method of assigning a high score to an item which more increases the probability of relatively high-probability causes and more decreases the probability of relatively low-probability causes when an input corresponding to the item is obtained.
4 . The information processing device according to claim 2 ,
wherein the processor is configured to generate input procedure information indicating the order of the input variables, based on the input procedure generation method.
5 . The information processing device according to claim 1 ,
wherein the processor is configured to:
specify an output condition linked to the important KPI information,
specify the cause linked to the output condition, and
set information corresponding to the cause as an output variable output from the AI engine.
6 . The information processing device according to claim 1 ,
wherein an important KPI information is designated for each of a plurality of user.
7 . An input control method of performing an input from an information processing device to an AI engine having a cause and an item as probability variables and represented by a network defining conditional dependency between the cause and the item, which executes steps comprising:
important KPI reception processing of receiving important KPI information designated by a user;
variable selection processing of selecting an input variable to be input to the AI engine, based on the important KPI information;
storing in association with each other, in the storage section, a combination of a plurality of input conditions and a plurality of input procedure generation methods which are rules for ordering the items;
specifying the plurality input conditions associated with the important KPI information;
specifying a plurality of the items associated with the priority KPI information;
selecting the input procedure generation method based on the input conditions;
generating input procedure information for determining an order of the specified items based on the input procedure generation method;
transmitting the input procedure information via an external terminal;
receiving an item corresponding to the transmission and an answer to the item via the external terminal; and
calculating a probability of an output variable that is the cause using the AI engine with the item and the answer to the item as input variables of the AI engine.
8 . The input control method according to claim 7 ,
wherein the processor is configured to select, as the input procedure generation method one of:
a cause identification-emphasized input procedure generation method of assigning a high score to an item which more increases the probability of a high-probability cause when an input corresponding to the item is obtained, and
a cause isolation-emphasized input procedure generation method of assigning a high score to an item which more increases the probability of relatively high-probability causes and more decreases the probability of relatively low-probability causes when an input corresponding to the item is obtained.
9 . The input control method according to claim 7 ,
wherein an important KPI information is designated for each of a plurality of users.
10 . A non-transitory computer readable medium with an executable program stored thereon to cause an information processing device to perform an input from the information processing device to an AI engine having a cause and an item as probability variables and represented by a network defining conditional dependency between the cause and the item, and to further cause the information processing device to execute steps comprising:
important KPI reception processing of receiving important KPI information designated by a user;
variable selection processing of selecting an input variable to be input to the AI engine, based on the important KPI information;
storing in association with each other, in the storage section, a combination of a plurality of input conditions and a plurality of input procedure generation methods which are rules for ordering the items;
specifying the plurality input conditions associated with the important KPI information;
specifying a plurality of the items associated with the priority KPI information;
selecting the input procedure generation method based on the input conditions;
generating input procedure information for determining an order of the specified items based on the input procedure generation method;
transmitting the input procedure information via an external terminal;
receiving an item corresponding to the transmission and an answer to the item via the external terminal; and
calculating a probability of an output variable that is the cause using the AI engine with the item and the answer to the item as input variables of the AI engine.